A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 194 2

Evolution of Thermal Energy Storage for Cooling ApplicationsBY BRUCE B. LINDSAY, P.E., MEMBER ASHRAE; JOHN S. ANDREPONT, MEMBER ASHRAE

CONTRIBUTORS: CHUCK DORGAN, PH.D., MEMBER ASHRAE; MARK MACCRACKEN, MEMBER ASHRAE; LARRY MARKEL, MEMBER ASHRAE; DOUGLAS REINDL, PH.D., P.E., MEMBER ASHRAE; PETER TURNBULL; VERLE WILLIAMS, MEMBER ASHRAE

COORDINATOR: GEOFF BARES, ASSOCIATE MEMBER ASHRAE

Thermal energy storage (TES) for cooling can be traced to ancient Greece and Rome where snow was transported from distant mountains to cool drinks and for bathing water for the wealthy. It flourished in the mid-1800s in North America where block ice was cut from frozen lakes and shipped south in insulated rail cars for food preserva-tion and health-care facilities.

Block ice (Photo 1) was initially used for cooling in

theaters, where a 4 ft × 4 ft × 2 ft (1.22 m × 1.22 m × 0.61

m) slab of ice would weigh 1 ton (907 kg) and, with the

heat of fusion of 144 Btu/lb (907 kg), would provide

12,000 Btu/h (334 kJ/kg) over 24 hours, hence our

industry terminology. The dairy industry was the first to

employ mechanical ice making to rapidly cool milk and

frugal farmers would build ice all night long to reduce

equipment size and cost.

First Generation of Thermal Energy StorageCooling of commercial office buildings became

widespread after World War II, and its availability

contributed to the rapid population growth in the

southern and western United States. Window units, split

DX, rooftop packages, and central chiller plants filled

their respective niches. Utilities recognized that air con-

ditioning was contributing to peak demand growth and

initially promoted conventional air conditioning and

refrigeration to increase revenues. Since the generat-

ing plants were underused at night, the utilities looked

for ways to build additional off-peak load. Thermal

energy storage for cooling office buildings and factories

was embraced and many demonstration projects were

initiated. However, due to the regulatory environment,

these programs had to be “revenue neutral” and not

125CELEBRATING YEARS

Bruce B. Lindsay, P.E., is manager, energy & resource conservation for Brevard Public Schools. John S. Andrepont is founder and president of The Cool Solutions Company, Lisle, Ill. Chuck Dorgan, Ph.D., is professor emeritus, University of Wisconsin – Madison; Mark MacCracken is president of CALMAC, Fair Lawn, N.J.; Larry Markel is technical staff consultant, Oak Ridge National Laboratory, Oak Ridge, Tenn.; Douglas Reindl, Ph.D., P.E., is a professor at the University of Wisconsin – Madison; Peter Turnbull is principal, commercial buildings and zero net energy program manager at Pacific Gas and Electric Company, San Francisco; Verle Williams is president, Utility Services Unlimited, Inc., Escondido, Calif.; Geoff Bares is director of engineering, Enwave Chicago.

This article was published in ASHRAE Journal, October 2019. Copyright 2019 ASHRAE. Posted at www.ashrae.org. This article may not be copied and/or distributed electronically or in paper form without permission of ASHRAE. For more information about ASHRAE Journal, visit www.ashrae.org.

O CT O B E R 2 0 19 a s h r a e . o r g A S H R A E J O U R N A L 4 3

subsidize one rate class at the expense of another. This

meant that the TES project would shift its peak kWh at

$0.05/kWh to off-peak at $0.025/kWh, but it had to con-

sume twice as much energy at night. There was no cus-

tomer savings, but the utility could provide an incentive

to balance its load profile. These technical requirements

favored ice storage and particularly “ice harvesting”

systems (see later section, “Cool TES Technology Family

Tree.”)

The equipment manufacturers, utilities, and engi-

neering firms saw a value in design guides and techni-

cal information. Sizing tanks, estimating weekly load

profiles, and performance of ice slurries was neither

well understood nor documented. ASHRAE established

Technical Committee (TC) 6.9, Thermal Storage, in 1981.

TES activity had previously been part of TC 6.1, Heat

Pumps; but the research focus did not align across two

very different markets.

Electric Utility Impacts and Second Generation of TESIn 1979, the electric industry fundamentally changed

when the Three Mile Island (TMI) nuclear plant expe-

rienced an incident. Previously, nuclear power was

supposed to be so inexpensive, it would not have to be

metered. Safety concerns and public opposition rapidly

increased the cost of nuclear plants with utilities pushed

to the brink of bankruptcy. Utilities actually looked at

ways to reduce growth to avoid additional construction

costs. Demand-side management (DSM) was embraced

to address peak electric demand. As a secondary result of

the TMI accident, the Electric Power Research Institute

(EPRI)—established by the utility industry to conduct

collaborative research—began to take a closer look at

advanced technologies for using electricity. As part of its

research into DSM, EPRI invested in a portfolio of tech-

nology developments involving TES.

The new generation of TES systems had a new focus—

reduce peak demand. The systems did not have to be

revenue-neutral, which had mandated less efficient

solutions such as ice harvesting. Simple ice tanks and

chilled water storage were allowable. Chilled water

storage was seen as the preferred technology by the

chiller manufacturers as their existing product lines

required no changes; but the challenge was to avoid

mixing the supply and return chilled water to maxi-

mize capacity and maintain cool supply temperature.

The TES industry experimented with various designs

(see later section, “Cool TES Technology Family Tree”)

before settling on thermal stratification as the sim-

plest strategy. Extensive research was conducted at

the University of New Mexico and separately by tank

manufacturers to develop theories affecting diffuser

designs to create and maintain stratification. EPRI

funded studies and ASHRAE TC 6.9 produced the

Design Guide for Cool Thermal Storage.

Ice storage tanks were also further developed in the

early 1980s. These included ice-on-coil internal melt,

ice-on-coil external melt, and encapsulated ice TES,

as well as ice slurries and other phase change materi-

als (PCMs), all described in the later section, “Cool TES

Technology Family Tree.”

A New ApproachThe perceived energy penalty for ice-making was

an impediment, especially as global warming con-

cerns were emerging. The TES industry developed a

novel alternative that radically departed from con-

ventional HVAC design. Air is distributed through

ductwork in most buildings at 55°F (12.8°C) in order

to achieve ASHRAE Standard 55 comfort conditions of

76°F (24.4°C) and 50%RH. Chilled water is typically sup-

plied to air-handling units at 44°F (6.7°C). An ice plant

can provide chilled water temperatures at nominal

32°F to 36°F (0 to 2.2°C), and its larger Delta T is wasted.

However, if the air-distribution system is designed for a

much lower supply temperature of 45°F (7.2°C), the air-

flow can be cut in half for the same cooling capacity. Fan

and duct size are reduced, offsetting the cost of the ice

PHOTO 1 Block ice farming. (courtesy of Wisconsin Historical Society)

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 194 4

storage system. EPRI conducted studies and produced

case studies documenting the energy savings and first

cost savings of cold air distribution (CAD) systems. EPRI

and Florida Power & Light (FP&L) funded one CAD/ice

demonstration project at Brevard Schools.

EPRI was involved extensively in developing, evaluating,

and promoting these different cool thermal energy storage

technologies. It pursued a portfolio management approach,

recognizing that there was not a one size fits all solution.

One philosophical change was the use of partial storage to

reduce first cost and limit the plant from bringing spare

chillers on-line in future years. EPRI worked closely with

ASHRAE TC 6.9 to disseminate information and fine-tune

its research agenda (Photo 2). TC 6.9 had quickly grown to

one of the largest technical committees due to its diverse

membership of manufacturers, utility personnel, consult-

ing engineers, academics, and facility managers seeking

detailed performance data and design guidance. The TC

was very active in research, handbook, and programs,

sponsoring many forums, seminars, and papers.

EPRI recognized that it needed to centralize technical

assistance to respond to its members’ requests for train-

ing and support. In 1991, EPRI established the Thermal

Storage Applications Research Center (TSARC) at the

University of Wisconsin-Madison. The UW was selected

due to its Engineering Professional Development program

and its HVAC workshops on TES, HVAC controls, and

cogeneration. Professor Charles E. Dorgan, Ph.D., P.E. was

the director and quickly added staff to fulfill the training,

communication, and technical assistance needs. TSARC

conducted workshops throughout the United States for

consulting engineers, sharing design experience and

practical insight. TSARC established an advisory commit-

tee to guide its research programs and focus on technical

resources. The committee included major utilities, manu-

facturers, researchers, and regulatory personnel.

The establishment of TSARC marks the height of utility

industry promotion of cool TES. Utilities were offering

incentives of up to $500 per kW avoided, sponsoring

training workshops, and installing TES systems in their

offices and operations facilities. That lit a fuse and TES

systems were exploding throughout the United States.

EPRI and TSARC also continued research into ice slurry

TES and PCM TES.

Cool TES Technology “Family Tree”Cool TES technologies can be divided into two main

branches: those storing energy as a change in phase

(latent heat systems) and those storing energy as a

change in temperature (sensible heat systems).

Most latent heat TES systems employ water-ice as the

phase change medium, though a minority of others have

used other phase change materials (PCMs). Primary

benefits are high energy density (low volume per stored

ton-hour) and modularity, while drawbacks include

complexity, the need for heat transfer to charge and dis-

charge TES, high energy consumption due to low temp

chiller operation, and little economy-of-scale. Ice TES

has taken the form of a variety of configurations, each

discussed below.

Latent Heat TESIce Harvester TES

Many manufacturers, including Paul Mueller, Turbo

Refrigerating, Henry Vogt, and others offered ice har-

vesting systems that produced sheet ice, tube ice, or

ice cubes on vertical heat transfer surfaces. A periodic

defrost cycle was used to shuck the ice into storage tanks

below, where water was later circulated through the

piles of ice to meet cooling loads. Due to the high unit

cost ($/ton) of the ice harvesters, they were generally

configured as weekly-cycles in which ice was produced

all weekend and each weeknight, and melted each week-

day; thus, the ice makers were smaller and less expen-

sive, while the ice tanks became several times larger. The

complexity of these “dynamic” ice systems led to O&M

issues, with simpler “static” ice options coming to domi-

nate the Ice TES market.

Ice-on-Coil—Internal MeltThis approach generally takes one of two forms. In

PHOTO 2 TC 6.9 Programs Chuck Dorgan seminar. (courtesy U of Wisconsin-Madison)

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 194 6

the first version, as long practiced by BAC, Evapco, and

others for modules of roughly 500 to 1,500 ton-hours

(1.8 to 5.3 MWh), a rectangular storage tank flooded with

water contains a serpentine coil of metal pipe through

which water-glycol is circulated. Cold glycol from chill-

ers serves to chill the pipes, forming ice on the pipe

exterior; later warm glycol from cooling loads serves

to melt the ice, from the inside-out. In the second ver-

sion, as long practiced by Calmac and Fafco for modules

of roughly 150 to 200 ton-hrs (0.5 to 0.8 MWh, a plastic

tank flooded with water contains small plastic tubing (or

other heat transfer means) through which water-glycol

is circulated. Again, cold glycol from chillers serves to

chill the pipes, forming ice on the pipe exterior; later

warm glycol from cooling loads serves to melt the ice,

from the inside-out. In some cases, nearly 100% of the

water can be converted to ice.

Ice-on-Coil—External MeltThis is a modified approach of the first type of inter-

nal melt ice-on-coil equipment described above. In

early examples, practiced by BAC, Evapco, and oth-

ers for modules of roughly 500 to 1,500 ton-hrs (1.8 to

5.3 MWh), a rectangular storage tank flooded with

water contains a serpentine coil of metal pipe through

which refrigerant is circulated and vaporized, forming

ice on the pipe exterior. In later examples, water-glycol

is circulated within the pipes to form ice. In both cases,

warm water from cooling loads flows through the tank

to melt the ice via direct contact, from outside-in. This

permits a cooler supply water temperature to cooling

loads and is especially applicable to district cooling

applications where the cooler supply temp can reduce

distribution pipe size and cost. In all cases, much less

than 100% of the water can be converted to ice, as

channels between the ice must be kept free for water

flow during ice melting.

Encapsulated Ice TESSimilar to ice storage tanks, encapsulated ice systems

were developed. Small plastic balls or lenses were filled

with water and placed in a storage container. A water/gly-

col solution (or other secondary fluid) would flow around

the balls or lenses and freeze the water. Later, warm

return fluid from the cooling loads would flow to melt the

ice and provide chilled water through a heat exchanger.

Manufacturers included Cristopia, Cryogel, Reaction Ice,

and others. Some manufacturers offered phase change

materials (PCMs) other than water to raise the freezing

temperature, allowing conventional chillers to be used

and minimizing the energy penalty of making ice.

Ice Slurry TESThe production of pumpable ice slurries has long been

explored, as the potential benefits from reductions in

pipe and pump sizes and pump energy are quite sub-

stantial. Various attempts have been made, including

using scraped-surface ice-makers (in Canada), Liquid

Ice by CB&I in the late 1970s/early 1980s, and Slippery

Ice by EPRI in the late 1980s, the latter two using falling-

film HXs with electropolished stainless steel tubes and

mylar heat transfer surfaces, respectively; however,

limitations of low heat flux and high unit cost ended fur-

ther development. A system operating at the triple-point

of water (with all three phases: solid, liquid and vapor

in equilibrium) was developed by IDE Technologies in

Israel in the 1970s, and has the benefit of direct contact

between the refrigerant (water vapor) and the water-

ice, operating at 32°F (0°C) and a deep vacuum; subse-

quently, it has found occasional niche applications in

the cooling of very deep gold mines in South Africa, as

a heat pump for district heating in Scandinavia, and as

a means of warm-weather snow-making for ski resorts.

But pumping ice slurries continues to be a largely unre-

alized “holy grail” due to issues of clogging in real-world

piping systems.

Beyond the latent heat systems in which water-ice is

the dominant choice for the storage medium, some have

employed other phase change materials (PCMs).

Paraffin Wax SlurriesIn the early 1980s, Drexel University performed

ASHRAE-funded research of pumpable slurries of paraf-

fin wax in water. Small globules of wax (with a freezing

point between typical CHW supply and return tem-

peratures) were suspended in water, becoming liquid at

the cooling load heat exchangers and returning to solid

globules at the chiller; however, issues with the wax

clogging heat exchangers ended further development.

Passive PCM TESIn the 1980s, paraffin waxes embedded in building

wallboards were also explored as passive PCM TES; how-

ever, issues of flammability ended further development.

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 194 8

Eutectic SaltsAlso, in the 1980s, a eutectic salt PCM was marketed

commercially by Transphase with a freezing point of

47 °F (8°C). It promised key benefits of Ice TES (having

high energy density) and CHW TES (being charged with

conventional CHW supply temperatures of 40°F to 42°F

or 4°C to 6°C); however, during the discharge (melt)

cycle of the encapsulated PCM, CHW supply tempera-

tures quickly rose well above the phase change tem-

perature, thus limiting the applications to partial shift

TES with chillers necessarily running downstream of

the TES. This, combined with system costs and material

environmental issues, ended the commercial applica-

tions after a number of years.

Triple Point Carbon DioxideIn the late 1980s, CB&I and Liquid Carbonic devel-

oped and demonstrated a very low-temperature TES

phase change technology using carbon dioxide at its

triple point (i.e., where all three phases, solid, liquid

and vapor, are in equilibrium). Deemed SECO2 (for

Stored Energy in Carbon Dioxide) it used CO2 vapor as

its refrigerant to achieve a liquid-solid phase change at

–70°F (– 57°C) and 60 psig (4.1 atmg).

Sensible Heat TESMost sensible heat TES systems employ water as the

storage medium, though a minority of others have used

other low temperature fluids (LTFs). Primary benefits are

simplicity, energy efficiency, and high economy-of-scale,

while drawbacks include low energy density (high volume

per stored ton-hr). Chilled water (CHW) TES has taken the

form of a variety of configurations, each discussed below,

with different means for maintaining necessary separa-

tion between cool supply water and warm return water.

Empty Tank Method TESIn its simplest configuration, the “empty tank” method

employs just two tanks: one to hold the cool supply water

and one to hold the warm return water; this keeps the

two temperature zones separate, but requires a 100%

increase in tank volume versus the water volume. To

minimize this excess tank volume and cost, systems were

sometimes configured with more than two tanks, some-

times as many as six, ten, or more tanks, always with all

but one of the tanks being adequate to hold the full water

volume; in this manner cool supply water and warm

return water could still always be kept separate, as once

a given “empty tank” had been filled, another tank had

become empty and available to receive the subsequent

flow of water. The larger the number of tanks, the smaller

the required excess tank volume; however, smaller

tanks have a higher unit capital cost; and each additional

tank requires full size inlet/outlet piping and valves to

handle the peak charge and discharge flows. Installations

occurred periodically in the 1970s and 1980s; but perhaps

the last significant example was in 1990 at Arizona State

University, where five (plus one) underground tank com-

partments contained 5.5 million gallons (20,800 m3) for

54,000 ton-hrs (190 MWh) of TES.

Labyrinth Tank TESLabyrinth tanks use a multitude of compartments in a

horizontal layout with connections such that water flows

in a long circuitous path from cell to cell, in one direction

during charging and reversed during discharging. Dating

to at least the 1970s, this technique was seen primarily in

Japan in the sub-basements of high-rise buildings where

the foundations already employed an “egg-crate” (hori-

zontal grid) construction for reasons of seismic design.

However, some temperature mixing would occur.

Baffle-and-Weir TESAlso dating to the 1970s, baffle-and-weir tanks were

occasionally used, where internal walls separated com-

partments within a tank, with water flowing over a wall

to one compartment then under a wall to the next, and

so on, with the flow reversed between charging and dis-

charging. However, again, temperature mixing would

occur.

Membrane or Diaphragm Separation TESDuring the 1980s, primarily in Ontario, often based

on designs by Robert (Bob) Tamblyn of Engineering

Interface, tanks used internal flexible membranes to

separate the upper warm zone from the lower cool

zone; however, operational problems such as fail-

ures of the membranes, blocked pump suctions, and

convection causing temperature mixing, were com-

mon issues. The first district cooling utility, begun

by Hartford Steam in Connecticut in 1964, added a

20,000 ton-hr (70 MWh) CHW TES tank in 1985 using

a rigid horizontal diaphragm designed to move up

and down during charging and discharging; however,

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 195 0

the diaphragm quickly jammed and failed, resulting

in the tank being converted to thermally stratified

TES, as it has performed for the past 34 years.

Thermally Stratified TESThermal stratification relies on the density difference

between the more dense, cool supply water and the less

dense, warm return water to create and maintain sepa-

ration of the temperature zones with no physical barrier;

internal flow diffusers are required to slow the inlet and

outlet flows to avoid mixing. Seminal laboratory testing

was conducted by Professor Maurice “Bud” Wildin of

the University of New Mexico in the late 1970s and early

1980s, while Professor William Bahnfleth,Ph.D., P.E.,

of the Pennsylvania State University subsequently con-

ducted full-scale field testing and analysis, all of which

contributed to past and current information in the

thermal storage chapter of the ASHRAE Handbook. Just as

the was case for the other configurations of CHW TES,

thermally stratified CHW TES was occasionally seen in

the 1970s and early 1980s as “one-off” designs by consul-

tants for particular applications; common among these

were automotive plants where firewater storage tanks

were required, and for which insulation and flow dif-

fusers were designed and specified by Gordon Holness

at Albert Kahn Associates or William Harrison at Giffels

Associates among others, to place these water tanks

into dual-service as TES. At the start of the 1980s, CB&I

(Chicago Bridge & Iron, now McDermott), which had

built a number of the one-off tanks, developed “turn-

key” design-built CHW TES tanks, for which owners and

engineers could merely specify TES capacity, operating

temperatures, and peak flow rates, with the complete

design and thermal performance guarantees provided

by the tank supplier. This procurement approach (simi-

lar to that used for chillers, cooling towers, or other key

CHW system components) led to a dramatic increase

in the use of CHW TES, with CB&I having subsequently

executed hundreds of such installations with steel tanks,

and other tank builders following suit over time, includ-

ing Natgun (now DN Tanks) being an early entrant with

concrete tanks, preferred especially for in-ground tanks.

Beyond the sensible heat systems in which water is the

dominant choice for the storage medium, some have

employed other low temperature fluids (LTFs). Aqueous

calcium chloride brines were occasionally used for

low temperature applications; however, chlorides are

notoriously corrosive, and once environmental issues

eliminated use of the only viable corrosion inhibitor in

the early 1980s, this option became economically much

less practical. D-Limonene is a hydrocarbon (C10H16) with

a freezing point of –102°F (–74°C) and can thus be used as

a fluid storage medium for very low temperature (gen-

erally specialized and rare) applications. One example

dating to the 1960s or 1970s was a large aeronautical test

facility at Eglin Air Force Base in Florida; a hangar-sized

test chamber required that a massive make-up ambient

airflow be cooled to about –40°F (–40°C) but for only a

short test period. The solution was a staged TES system

of a two-tank “empty tank” method of calcium chloride

brine, in series with a two-tank “empty tank” method of

D-Limonene in two spherical pressure vessels.

Aqueous Sodium Nitrite/Nitrate TESOne particular version of low temperature fluid (LTF)

TES uses a patented, thermally stratified solution of

sodium nitrite and sodium nitrate in water, marketed as

SoCool fluid, developed by Trigen Energy Corporation,

with the patents later transferred to CB&I. The chemi-

cals not only lower the freezing point versus that of pure

water, they also lower the temperature at which maxi-

mum density occurs. This allows thermal stratification

below the limit for pure water which occurs at about

39.4°F (4.1°C). It also increases the supply-to-return

Delta T in TES, thus reducing the volume per ton-hr

versus that of conventional CHW TES. First employed in

1994, over the next 15 years there were a dozen installa-

tions totaling nearly 300,000 ton-hrs (1055 MWh), with

supply temperatures ranging from 29°F to 36°F (–2°C to

+2°C). Applications include Chicago’s McCormick Place

convention facilities (1994), DFW International Airport

(2002), and Princeton University (2005).

Pioneering Utility ProgramsTU Electric was experiencing incredible growth in and

around Dallas in the late 1980s, and needed to manage its

peak demand (Photo 3). It offered customer incentives of

$250/kW for the first 500 kW shifted to off peak and $125/

kW thereafter. Texas Instruments (TI) installed a 2.7 mil-

lion gallon (10,200 m3) stratified CHW TES tank in 1990 to

supplement its 4,200-ton (14 771 kW) chiller plant at its 1.1

million square foot (100 000 m2) electronics manufactur-

ing facility. TI invested $1.6 million for 24,500 ton-hours

(86 MWh). Two more CHW TES installations followed at

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 195 2

other TI facilities.

The 40-story, 1.5 million ft2

(140 000 m2) convention hotel in San

Francisco included 1,500 guest rooms

and 125,000 ft2 (11 600 m2) of meet-

ing space. It had a peak electric load

of 3,764 kW and a design cooling load

of 920 tons (3236 kW). The chilled

water plant consisted of two 900-ton

(3165 kW) chillers. In 1993, the hotel

was retrofitted with a supplemental

180-ton (633 kW) chiller (108 tons

in ice-making mode) and six 500

ton-hour (1.8 MWh) ice-on-coil TES

tanks (Photo 4). The ice TES plant was

operated to take advantage of Pacific

Gas & Electric real time pricing (RTP),

consisting of a base $0.04/kWh,

a “transmission and distribution

adder” up to $0.30/kWh, and a “gen-

eration and bulk transmission” adder

up to $0.80/kWh.

Florida Power & Light targeted the

k–12 school market to promote cool

TES in the early 1980s. It success-

fully convinced the School Board of

Brevard County to adopt ice stor-

age and cold air distribution (CAD).

The district wanted to provide

better humidity control and also

reduce the size and cost of ductwork

and fans. The initial systems were

Mueller ice harvester TES. These

were eventually converted to ice-

on-coil tanks. FPL rebates funded 20

ice plants serving 21 schools, total-

ing 10,640 tons (37 419 kW) of peak

reduction and over $60 million.

There were 17 CALMAC and 3 FAFCO

tank farms. The systems performed

well, taking advantage of a seasonal

time-of-use demand rate that only

charged demand charges between 3

and 6 p.m., Mon-Fri, for four sum-

mer months. All the schools were

closed at 3 p.m., but the ice storage

schools could continue operations

without a penalty. In 2008, tax rev-

enues fell drastically and the district

was forced to implement budget

cuts. Maintenance was neglected.

Valves, actuators, and sensors failed;

but there were not sufficient funds

for repairs/replacements. In 2014,

the taxpayers agreed to a massive

facility renewal program stretched

over six years. Most of the chillers

have been or will be replaced, and

the building automation systems

upgraded, but the ice storage tanks

were deemed serviceable. So far,

only one school replaced its tanks, in

March 2019 (Photo 5).

New Challenges for the HVAC Industry and Opportunities

The Clean Air Act of 1996 marked

another milestone for the TES

industry. The phaseout of CFC

refrigerants put the focus on chilled

water plants throughout the US.

PHOTO 4 San Francisco Marriott Hotel Ice TES. (courtesy PGE)

PHOTO 5 Ice Tank Installation at Brevard Schools, 2019. (courtesy Brevard Schools)

PHOTO 3 Texas Instruments CHW TES. (courtesy DN Tanks)

Facility managers were tasked with

containing leaks, evaluating the

costs of converting or replacing

chillers, selecting future refriger-

ants, and considering other options.

TES was one option with several

different wrinkles. Many facili-

ties integrated TES to supplement

capacity while decommissioning

a chiller to salvage the CFC refrig-

erant for future use. In Chicago,

Commonwealth Edison’s affiliate

(Northwind Chicago) built a district

cooling (DC) system using a massive

ice storage system (Photo 6). The ice

made it feasible to distribute very

cold water throughout the city’s

downtown “Loop,” raising Delta T,

and reducing pipe diameter and

pumping energy. The DC system

aggressively marketed the service

to buildings in the Loop as a simple

way to sidestep the CFC phaseout

and outsource chiller operations.

Of course, Commonwealth Edison

was pleased to sell off-peak power

at night from its nuclear fleet on a

long-term contract. DC with TES was

quickly applied in other cities.

The DC system grew to five plants

(four of which have ice TES).

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 195 4

Plant 1, built in 1995, had 66,000

ton-hrs (232 MWh). Plant 2 has a

TES capacity of 125,000 ton-hrs

(440 MWh), using BAC ice-on-coil

modules, and came on-line in 1996

(Photo 7). Plant 3, built in 1997, had

97,000 ton-hrs (341 MWh). Plant

4, expanded in 1997, has 26,000

ton-hrs (91 MWh) dating from

c.1985 (when it served only the

Merchandise Mart, prior to the start

of the DC system). Plant 5 built in

2002, has no TES. All the ice TES are

BAC ice-on-coil units.

The large gray tank in the left cen-

ter of Photo 8 is the first ever appli-

cation of the thermally stratified

aqueous sodium nitrite/nitrate low

temp fluid (LTF) TES system.  The

TES tank was completed in 1994 by

CB&I and serves the district energy

system developed by Trigen-Peoples

(now owned and operated by the

City of Chicago).  It serves all 4 large

exhibit hall buildings comprising

McCormick Place, plus nearby hotel,

office, and internet server facilities. 

The 8.5-million gallon (32 200 m3)

tank stores 123,000 ton-hours (433

MWh) at LTF supply/return temps

of 30/54°F (–1.1/+12.2°C), capable of

discharge rates of up to 25,000 tons

(88 MW) (up to an 18.75 MW peak

electric load shift).  The cold supply

temp was chosen for two reasons:

• to maximize Delta T, thus mini-

mizing tank volume in this urban

setting and

• to provide low temp air-

distribution in much of the then

newly-constructed exhibit halls,

thus minimizing the size and costs

of (miles of) air-ducts and fans, and

minimizing fan power.

The low temp fluid also provides

all on-going water treatment neces-

sary for corrosion inhibition and

microbiological control.

The TECO (Thermal Energy

Corporation) district energy sys-

tem provides steam and chilled

water to 18 hospitals in the Texas

Medical Center in Houston (Photo

9).  The 8.8 million gallon (33 300

m3) (100 ft D × 150 ft H [30.5 m

D × 45.7 m H]) chilled water TES

tank was provided by CB&I in 2010. 

It provides 70,000 ton-hours (246

MWh) at CHWS/R temps of 40/53°F

(4.4/11.7°C).  The TES can shift peak

loads of up to 14,000 tons (10 MW

electric).  Due to excess nighttime

wind power on the Texas grid,

TECO is sometimes paid as much

as $0.10/kWh to consume power to

recharge the TES.  And on extreme

peak days, the TES allows TECO to

avoid real-time power costs as high

as several dollars per kWh.

Utility Transformation ContinuesAt the end of the 1990s, the US

prepared for Y2K and the electric

utility industry was transforming

into a de-regulated, competitive,

non-integrated bundle of different

companies. There were GENcos,

TRANScos, DIScos, and RETAILcos,

and they were merging and divest-

ing. Incentives and support for cool

TES made a left turn. Why would a

local distribution company pay an

incentive to avoid building another

organization’s power plant? Power

plants were cheap and plentiful.

The retailers wanted to keep things

simple and sold electricity at one

price, with no on-peak, off-peak, or

PHOTO 7 Northwind Chicago Plant #2 under construction with external melt ice-on-Coil units. (courtesy Enwind Chicago)

PHOTO 6 Northwind Ice Storage Plant, Chicago. (courtesy Enwave Chicago)

PHOTO 8 Trigen-Peoples McCormick Place Stratified Low Temp Fluid (LTF) TES, Chicago. (courtesy Chicago Bridge & Iron Co.)

PHOTO 9 Thermal Energy Corporation (TECO), District Energy System, Houston, using stratified CHW TES. (courtesy TECO)

ASHRAE — CELEBRATING 125 YEARS

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 195 6

demand charges. It had become very difficult to justify a

cool TES investment with these uncertainties. EPRI shut-

tered its cool TES program due to a lack of funding. Then,

the ENRON scandal occurred in October 2001. California

power markets had gone insane.

Several events and market changes occurred in the

following years. Natural gas prices spiked in 2005 after

Hurricane Katrina dismantled the offshore piping net-

work in the Gulf. That rapidly reversed as fracking was

able to produce vast quantities of inexpensive natural gas

A Cool TES Hall of FameEight co-authors and contributors for this article nominated 27 individuals, in six categories, for an informal Cool

TES Hall of Fame, and then voted based on individuals having provided a unique or pioneering contribution, which has proven to have had staying power over the years. These 14 inductees, listed alphabetically by category, each received between 50% and 88% votes.

AcademicsWilliam (Bill) Bahnfleth, Pennsylvania State

University. He conducted field testing and analysis related to thermally stratified CHW TES, reflected in the ASHRAE Handbook and Design Guide. (50% of voters)

Charles “Chuck” Dorgan, The University of Wisconsin-Madison HVAC&R Center. He analyzed and championed the design and economics of ice TES and cold air distribu-tion. (88%)

Maurice “Bud” Wildin, The University of New Mexico. He performed seminal lab-scale testing of thermally stratified CHW TES, reflected in the ASHRAE Handbook and Design Guide. (63%)

ConsultantsIan Mackie, Mackie Associates. He designed diffusers

for thermally stratified CHW TES and helped engineer the first TES for utility turbine inlet cooling in 1991 (using ice harvesters). (63%)

Robert (Bob) Tamblyn, Engineering Interface Ltd. He developed and applied CHW Membrane Tank TES for numerous applications in Ontario in the 1980s. (50%)

Verle Williams, Utility Services Unlimited. He designed and applied Ice and CHW TES, each uniquely chosen and configured to best suit a variety of early appli-cations in the Western United States. (69%)

Equipment SuppliersJohn Andrepont, CB&I (Chicago Bridge & Iron Co.). He

pioneered and championed performance-guaranteed, thermally stratified CHW (and LTF) TES in the 1980s and early 1990s, leading to >250 projects and >4.5 million ton-hrs, later using TES as an owner, and then consulting for >100 TES projects. (88%)

Calvin (Cal) MacCracken, Calmac Manufacturing. The Babe Ruth of TES, he was an inventor (~70 patents) and entrepreneur who developed small (~150 ton-hr) internal-melt ice-on-coil modules, which have ultimately

been applied in over 4,000 projects in 60 countries, and total 6 million ton-hrs. (88%)

William (Bill) McCloskey, BAC (Baltimore Aircoil Company). He led the marketing and sales effort for large (~1,500 ton-hr) internal and external-melt, ice-on-coil TES, including urban district cooling systems in the United States, Canada and East Asia, ultimately applied in >3,000 projects, totaling 7 million ton-hrs. (63%)

OwnersDonald (Don) Fiorino, Texas Instruments. He imple-

mented and documented the use of TES, installing multi-million gallon thermally stratified CHW TES at several TI facilities in the early 1990s. (50%)

Edward (Ed) Knipe, California State University System. He advocated and oversaw the use of thermally stratified CHW TES on a dozen CSU campuses in the 1990s, ulti-mately totaling >300,000 ton-hrs. (50%)

UtilitiesLoren McCannon, San Diego Gas & Electric and ITSAC

(Int’l Thermal Storage Advisory Council). He oversaw one of the first utility TES incentive programs and edited/pub-lished the ITSAC newsletter. (63%)

John Nix, Florida Power & Light. He oversaw one of the longest lasting and most successful utility TES incen-tive programs, ultimately achieving ~400 projects and ~140,000 tons of peak load reduction. He also led devel-opment of an ASHRAE standard for TES testing. (75%)

OthersRonald (Ron) Wendland, EPRI (Electric Power

Research Institute). He was a chief advocate and central clearinghouse for all things TES-related for the utility industry and adeptly “herded the cats” comprising the many diverse TES equipment suppliers active during the 1980s and early 1990s. (63%)

ASHRAE — CELEBRATING 125 YEARS

O CT O B E R 2 0 19 a s h r a e . o r g A S H R A E J O U R N A L 57

PHOTO 10 LES Peaking Plant with Ice TES for Inlet Air Pre-Cooling. (courtesy LES)

and the utility industry discovered

the allure of combustion turbines

(CTs). The only problem with CTs is

that they are rated at 59°F (138.2°C)

and output de-rates dramatically on

hot summer days when it is needed

most. That generated a new applica-

tion for cool TES. Chilled water or ice

can pre-cool the inlet combustion air

and regain the full output of the CT.

TC 6.9 sponsored programs on this

new opportunity and development of

the Combustion Turbine Inlet Air Cooling

Systems Design Guide.

In 1991, the Lincoln Electric System

(LES) in Lincoln, Nebraska, with

assistance from EPRI, installed the

first utility Turbine Inlet Cooling

(TIC) system to use TES (Photo 10). A

weekly-cycle ice harvester TES sys-

tem was installed to pre-cool the CT

inlet air, recovering 20% of the lost,

hot weather output for 4 hours/day.

Other similar TES-TIC systems were

subsequently installed in Nebraska,

North Carolina, Wisconsin, and Saudi

Arabia during the 1990s; however, by

the late 1990s, CHW TES became the

dominant choice for TES-TIC, being

used in capacities of up to 700,000

ton-hrs (2462 MWh) at a single plant

in Saudi Arabia, for an increased hot

weather output of 30%.

In 2005, the Saudi Electricity

Company (SEC) in Riyadh, installed

a 192,800 ton-hr (678 MWh) daily-

cycle CHW TES system for TIC,

achieving a hot weather output

increase of 30% during six hours/

day. More and often larger CHW

TES-TIC installations have followed

in Saudi Arabia, Pennsylvania,

Virginia, Texas, New Mexico,

California, and Florida, serving both

simple cycle (peaking) CTs and CT

combined cycle plants.

20 Years Later (21st Century Postscript)

Cool TES is still a viable (and grow-

ing) technology today. Early cool stor-

age technology was inspired and sig-

nificantly funded by the electric utility

ASHRAE — CELEBRATING 125 YEARS

Advertisement formerly in this space.

A S H R A E J O U R N A L a s h r a e . o r g O CT O B E R 2 0 195 8

industry and supported technically by EPRI and ASHRAE.

The utility industry has changed. Cool TES has evolved. In

addition to US installations of ice TES numbering in the

many thousands, and CHW TES (on average much larger)

and numbering in the many hundreds, there have been

many and growing Ice and CHW TES applications around

the world, including Canada, Mexico, Brazil, Europe, East

Asia, Southeast Asia, the Middle East, South Africa, and

Australia. The cool TES industry, in many ways mature

and thriving, exhibits increasing numbers of applications

for traditional peak electric load management. Those

applications primarily use either small- to large-scale ice-

on-coil TES or large-scale thermally stratified CHW TES.

But additionally, there has been significant growth in new

types of TES applications or new markets, notably:

• Large District Cooling (DC) systems in East Asia (us-

ing CHW TES or mostly ice-on-coil TES),

• Very large DC systems in the Middle East (using ice-

on-coil TES or mostly CHW TES); note that DEWA (Dubai

Electricity & Water Authority) requires that new DC

systems employ TES for at least 20% of peak capacity,

• Small-scale “rooftop” TES (using ~30 to 50 ton-hr

[105 500 to 175 843 Wh] direct-refrigerant Ice TES units),

• Emergency back-up cooling for Mission Critical

Facilities, such as data centers (using mostly small- to

medium-scale CHW TES),

• Turbine Inlet Cooling (TIC) of Combustion Turbine

power plants (using mostly large- to very large-scale

CHW TES),

• The use of hot TES, as many district heating systems

are converted from steam to hot water (using thermally

stratified HW TES), and

• Support for increased deployment of intermittent

renewable power sources, such as wind and solar (using

all types of TES, as TES unit capital costs, life expectancy,

and other performance characteristics are often far

superior to those of batteries or other energy storage

technology options).

Also, new TES technologies have continued to be devel-

oped, some beginning (or targeted) to capture new or

niche market applications, e.g.:

• Ice slurry for cooling of human bodies and organs

ASHRAE — CELEBRATING 125 YEARS

Advertisement formerly in this space.

O CT O B E R 2 0 19 a s h r a e . o r g A S H R A E J O U R N A L 5 9

during surgery (using secondary aqueous fluid-ice pro-

duction), developed by the U.S. Department of Energy at

the Argonne National Laboratory,

• Ice slurry for warm weather production of snow for

ski-resorts (using triple-point production of water-ice), and

• Various non-water phase change material (PCM)

technologies, including bio-based (plant-based), non-

toxic, non-corrosive PCMs that change from solid-to-

solid phase or solid-to-“gel” as heat is added/removed.

Phase change temperatures can range from approxi-

mately –50°C to +170°C (–58°F to +338°F), from a number

of manufacturers. They can be used in passive or active

systems including in traditional internal melt ice-on-

coil tanks. The advantage of selecting a temperature to

suit specific applications, is often offset by high material

cost compared to that of water-ice or eutectic salts.

And although some TES installations are still driven

by varied utility incentives (such as demand charges,

time-of-use rates, real-time pricing, grid-wide coinci-

dent peaks, and even some cash incentives, with ConEd

of NYC offering $2,520/kW still!), many applications are

primarily justified by avoided capital investments in

conventional (non-TES) chiller plant capacity when that

capacity can be downsized through the use of TES, spe-

cifically during times of:

• New construction,

• Retrofit expansions, or

• Retirement/replacement of aging chiller plant

equipment.

The evolution of TES has been rich with technologies,

firms, and individuals. And the future of TES is bright

with the continuing promise of solutions for HVAC, the

electric grid, microgrids, and energy storage challenges.

References1. ASHRAE. 1993. ASHRAE Design Guide for Cool Thermal Storage, first

ed. Atlanta: ASHRAE.2. Kilpatrick, A.T., J.S. Elleson. 1996. Cold Air Distribution System

Design Guide. Atlanta: ASHRAE.

BibliographyASHRAE. 1999. Combustion Turbine Inlet Air Cooling Systems Design Guide.ASHRAE. 2013. District Cooling Guide, Chapter 6, Thermal Energy

Storage.2016 ASHRAE Handbook— HVAC Systems & Equipment, Chapter 51,

Thermal Storage.

ASHRAE — CELEBRATING 125 YEARS

Advertisement formerly in this space.